Effectively treating communication disorders after stroke requires accurate differential diagnosis. In some cases, however, we lack reliable, valid methods of differentially diagnosing disorders with similar behavioral profiles. This is particulary true of acquired apraxia of speech (AOS) and aphasia characterized by frequent occurrences of phonemic paraphasia (PP). The differential diagnosis of AOS and PP is challenging because both disorders result from left hemisphere stroke and share clinical characteristics. Therefore, the identification of characteristics that pattern uniquely to each disorder is important. One way in which to strengthen the current diagnostic process is to examine the validity of diagnostic criteria used to inform differential diagnosis. The current criteria proposed to differentiate AOS from PP include: 1) slow speech rate characterized by prolonged segment and intersegment durations, 2) sound distortions, 3) distorted sound substitutions, 4) prosodic abnormalities, and 5) relatively consistent errors in regard to error location and error type. Because it has been described as being relative, error consistency is one characteristic that holds promise for the differentiation of these two disorders. Error consistency refers to whether or not errors are relatively consistent from trial to trial in regard to the location of errors within a word (e.g., ord initial) and the type of errors produced (e.g., distortions vs. substitutions). Investigations comparing the nature of error consistency in AOS and PP have revealed conflicting results. However, these studies differ in important methodological areas, making it difficult to draw conclusions about the nature of error consistency in these two populations. Furthermore, the literature suggests that error consistency may be influenced by a number of variables, such as stimulus presentation condition, error rate, and severity of impairment. Therefore, the purpose of this proposal, is to further examine the nature of error consistency in a group of 10 individuals with AOS and concomitant aphasia and a group of 10 individuals with aphasia with PP. Specifically, this study will explore group differences in the consistency of error location and error type during the repetition of two-, three-, and five-syllable words. The influence of stimulu presentation condition, error rate, and severity of impairment on error consistency, within and between groups, will also be examined. Lastly, to further examine group differences in the types of errors produced the occurrence and frequency of specified error types will be compared. This study will make a significant contribution as the first step in a programmatic line of research focused on improving the differential diagnosis of AOS from its clinical neighbors by examining specific clinical characteristics and patterns of characteristics in individuals with acquired communication impairments resulting from left hemisphere stroke.